Servicing an access request using a plurality of replicas of data includes determining which of the replicas are available and selecting one of the replicas for servicing the access request based on which of the replicas are available. At least one other replica is indicated as not available in response to receiving a monitoring message indicating that the at least one other replica is subject to degraded performance. The monitoring messages are independent of the access request. The monitoring messages may be SNMP messages. Servicing an access request using a plurality of replicas of data may also include determining if each of the SNMP messages affects the availability of the replicas. The replicas may be selected in round robin fashion or randomly.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method of servicing an access request using a plurality of replicas of data, the method comprising: determining which of the replicas are available, wherein each of the replicas is included in a list of replicas, and wherein availability of each of the replicas is determined by a controller from the list of replicas, wherein the plurality of replicas include multiple instances of same data on different servers; and selecting one of the replicas for servicing the access request based on which of the replicas are determined as available by the controller using the list of replicas, wherein at least one other replica on the list of replicas is indicated as not available in response to receiving a particular monitoring message from the at least one other replica indicating that the at least one other replica is subject to degraded performance, and wherein monitoring messages received from the plurality of replicas are independent of the access request, wherein each of the monitoring messages is initiated by a corresponding one of the plurality of replicas, and wherein the at least one other replica is indicated in the list of replicas as unavailable based on receipt of the particular monitoring message from the at least one other replica independent of the controller performing an attempt to access or transmit a communication to the at least one other replica.
2. A method, according to claim 1 , wherein the monitoring messages are SNMP messages.
3. A method, according to claim 2 , further comprising: determining if each of the SNMP messages affect the availability of the replicas.
4. A method, according to claim 1 , wherein the replicas are selected in round robin fashion.
5. A method, according to claim 1 , wherein the replicas are selected randomly.
6. A method, according to claim 1 , further comprising: in response to the selected replica not performing the access request, the replica is indicated as not available.
7. A method, according to claim 6 , further comprising: retrying an access operation.
8. A non-transitory computer-readable medium storing computer software that, when executed by one or more processors of a computer system, causes the computer system to implement a method that services an access request using a plurality of replicas of data, the method comprising: determining which of the replicas are available, wherein each of the replicas is included in a list of replicas, and wherein availability of each of the replicas is determined, by a controller, from the list of replicas, wherein the plurality of replicas include multiple instances of same data on different servers; and selecting one of the replicas for servicing the access request based on which of the replicas are determined as available by the controller using the list of replicas, wherein at least one other replica on the list of replicas is indicated as not available in response to receiving a particular monitoring message from the at least one other replica indicating that the at least one other replica is subject to degraded performance, and wherein monitoring messages received from the plurality of replicas are independent of the access request, wherein each of the monitoring messages is initiated by a corresponding one of the plurality of replicas, and wherein the at least one other replica is indicated in the list of replicas as unavailable based on receipt of the particular monitoring message from the at least one other replica independent of the controller performing an attempt to access or transmit a communication to the at least one other replica.
9. A non-transitory computer-readable medium, according to claim 8 , wherein the monitoring messages are SNMP messages.
10. A non-transitory computer-readable medium, according to claim 9 , the method further comprising: determining if each of the SNMP messages affect the availability of the replicas.
11. A non-transitory computer-readable medium, according to claim 9 , the method further comprising: indicating that the replica is not available in response to the selected replica not performing the access request.
12. A non-transitory computer-readable medium, according to claim 9 , the method further comprising: retrying an access operation.
13. A non-transitory computer-readable medium, according to claim 8 , wherein the replicas are selected in round robin fashion.
14. A non-transitory computer-readable medium, according to claim 8 , wherein the replicas are selected randomly.
15. A data storage system, comprising: a plurality of clients, each having at least one hardware processor, that access files; and a plurality of interconnected servers, each having at least one hardware processor and being coupled to the clients, at least one of the servers performing the following processing: determining which of a plurality of replicas are available for servicing an access request, wherein each of the replicas is included in a list of replicas, and wherein availability of each of the replicas is determined, by the at least one server, from the list of replicas, wherein the plurality of replicas include multiple instances of same data on different servers, and selecting one of the replicas for servicing the access request based on which of the replicas on the list of replicas are available, wherein at least one other replica on the list of replicas is indicated as not available in response to receiving a particular monitoring message from the at least one other replica indicating that the at least one other replica is subject to degraded performance, and wherein monitoring messages received from one or more of the plurality of replicas are independent of the access request, wherein each of the monitoring messages is initiated by a corresponding one of the plurality of replicas, and wherein the at least one other replica is indicated in the list of replicas as unavailable based on receipt of the particular monitoring message from the at least one other replica independent of the controller performing an attempt to access or transmit a communication to the at least one other replica.
16. A data storage system, according to claim 15 , wherein the monitoring messages are SNMP messages.
17. A data storage system, according to claim 16 , wherein the at least one of the servers also determines if each of the SNMP messages affect the availability of the replicas.
18. A data storage system, according to claim 15 , wherein the replicas are selected in round robin fashion.
19. A data storage system, according to claim 15 , wherein the replicas are selected randomly.
20. A data storage system, according to claim 15 , wherein the at least one of the servers indicates that the replica is not available in response to the selected replica not performing the access request.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
June 30, 2010
October 28, 2014
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.